- Email: [email protected]
Risk factors associated with snake antivenom reaction and the role of skin test
Journal Pre-proof
Risk factors associated with snake antivenom reaction and the role of skin test
Po-Chun Chuang ConceptualizationFormal analysisData CurationWriting - Original DraftWriting - Review & Editing Kang-Wei Chang Data Curation , Fu-Jen Cheng ResourcesValidationSupervision , Meng-Huan Wu ResourcesValidationSupervision , Ming-Ta Tsai ResourcesValidation , Chao-Jui Li MethodologyFormal analysisResourcesWriting - Review & EditingSupervision PII: DOI: Reference:
S0001-706X(19)31445-7 https://doi.org/10.1016/j.actatropica.2019.105293 ACTROP 105293
To appear in:
Acta Tropica
Received date: Revised date: Accepted date:
18 October 2019 22 November 2019 6 December 2019
Please cite this article as: Po-Chun Chuang ConceptualizationFormal analysisData CurationWriting - Original DraftW Kang-Wei Chang Data Curation , Fu-Jen Cheng ResourcesValidationSupervision , Meng-Huan Wu ResourcesValidationSupervision , Ming-Ta Tsai ResourcesValidation , Chao-Jui Li MethodologyFormal analysisResourcesWriting - Review & EditingSupervision , Risk factors associated with snake antivenom reaction and the role of skin test, Acta Tropica (2019), doi: https://doi.org/10.1016/j.actatropica.2019.105293
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Published by Elsevier B.V.
Risk factors associated with snake antivenom reaction and the role of skin test
Po-Chun Chuang, Kang-Wei Chang, Fu-Jen Cheng, Meng-Huan Wu, Ming-Ta Tsai, and Chao-Jui Li*
Department of Emergency Medicine, Kaohsiung Chang Gung Memorial Hospital, Chang Gung University College of Medicine, No. 123, Dapi Rd., Niaosong Dist., Kaohsiung City 833, Taiwan (R.O.C.)
*Corresponding author: Chao-Jui Li, Department of Emergency Medicine, Kaohsiung Chang Gung Memorial Hospital, Chang Gung University College of Medicine, No. 123, Dapi Rd., Niaosong Dist., Kaohsiung City 833, Taiwan (R.O.C.) Tel.: +886-912-188-889; Fax: +886-7-7317123-8415; Email: [email protected]
Abstract Antivenom reactions are a common complication of snake antivenom. This study aimed to identify predicators of antivenom reaction and the involvement of antivenom skin test in antivenom reaction development. This retrospective cohort study was conducted in six medical institutions in Taiwan. Data were extracted from the Chang Gung Research Database (CGRD) from January 2006 to December 2016. The association between antivenom reaction and patient demographics, type and dose of antivenom, and skin test results was analyzed. The study enrolled 799 patients, including 219 who developed antivenom reactions. Compared to patients receiving both freeze-dried hemorrhagic (FH) and freeze-dried neurotoxic (FN) antivenom, those administered a single type had a lower antivenom reaction risk (adjusted odds ratios [aORs]: 0.5 and 0.4, 95% confidence interval [CI]: 0.35–0.74 and 0.24–0.69, FH and FN respectively). Patients administered a higher antivenom dose (≥ 5 vials) had higher antivenom reaction risk (aOR: 1.8, 95% CI: 1.23–2.76). A positive skin test result was also associated with antivenom reaction (aOR: 16.7, 95% CI: 5.42– 51.22). The skin test showed high specificity (98.5%, 95% CI: 97.49%–99.83%) but low sensitivity (17.5%, 95% CI: 10.74%–24.18%). The antivenom skin test should be abolished because of the extremely low sensitivity and possible misinterpretation of results because of the limitation of this examination.
Keywords: snakes, antivenins, allergy, antivenom reactions, skin test, Taiwan
1. Introduction Snakebite envenomation is a tropical disease that causes mortality and morbidity in more than 100,000 and over 400,000 people, respectively yearly.1-3 Taiwan is located in Southeast Asia and has approximately 23 species of snakes, including 6 common clinically significant species. 4 Snake venom toxicity can be divided into hemorrhagic including that caused by the green habu (Viridovipera stejnegeri), Taiwan habu (Protobothrops mucrosquamatus), and hundred-pacer (Deinagkistrodon acutus) snakes, and neurotoxic, which includes toxicity caused by the Taiwan banded krait (Bungarus multicinctus) and the Chinese cobra (Naja atra), while the Russell’s viper (Daboia siamensis) has mixed hemorrhagic and neurotoxic venom. Treatment of snake bites mainly involves the injection of horse immunoglobulin (Ig) antivenom. Horse Ig antivenoms produced by the Centers for Disease Control, R.O.C (Taiwan) are freeze-dried hemorrhagic (FH) bivalent antivenom against the hemorrhagic venom of V. stejnegeri and P. mucrosquamatus; freeze-dried neurotoxic (FN), a bivalent antivenom against the neurotoxic venom of B. multicinctus and N. atra; freeze-dried acutus (FA), an antivenom against the hemorrhagic venom of D. acutus; and freeze-dried D. russellii (FR), which was developed in 2008 against the mixed-type venom of D. siamensis.4-8 These antivenoms are vacuum freeze-dried F(abʹ)2 fragments from horses, which contain more than 1000 units of antivenom serum; and the excipient contains the preservative (Thiomersal).9 Combining F(abʹ)2 fragments, which are 100 kDa with a half-life of approximately 2–4 days, with snake venom antigens forms stable and easily cleared immune complexes.10
Snake antivenoms can cause reactions ranging from mild skin rash to life-threatening anaphylaxis.11 Early reactions occur soon after antivenom injection (often within 1 h)11, 12, whereas late (serum sickness) reactions could develop between 5 and 14 days.11, 13 Early reactions could be due to the combination of type I (IgE) hypersensitivity, immune complex aggregation, and activation of the complement.11, 14 Late (serum sickness) reactions are mediated by type III (IgG or IgM) hypersensitivity.14 To predict antivenom reactions, the current treatment principle and manufacturer instructions of antivenom in Taiwan suggest that a skin test should be conducted before administration.9,
15, 16
However, the effects of skin tests are unclear, especially because
clinically regardless of the results of the skin test, physicians should use antivenoms to treat snakebites.17, 18 A previous case series study reported that the skin test was not helpful in predicting antivenom reactions19, and its sensitivity and specificity were unclear. Furthermore, the World Health Organization (WHO) guideline suggests that an antivenom skin test should not be used.20 Therefore, this study aimed to analyze the skin test accuracy and risk factors associated with antivenom reaction.
2. Methods 2.1 Ethics This retrospective study was approved by the Chang Gung Medical Foundation Institutional Review Board (IRB No.: 201801929B0). All patient data used in the analyses were anonymized.
2.2 Study setting The data used were obtained from the largest healthcare system in Taiwan, the Chang Gung Memorial Hospital (CGMH), which receives 11.5% of the National Health Insurance budget according to government statistics. The Chang Gung Research Database (CGRD), which combines original medical records from six medical institutions, Keelung, Taipei, Linkou, Yunlin, Chiayi, and Kaohsiung branches located from northern to southern Taiwan, was used.
2.3 Patients All patients who experienced snake bites, visited the emergency department (ED), and received antivenom injections from January 2006 to December 2016 were included in the study. Patients who received any class of antihistamine (including H1 or H2 blockers), steroid (including systemic or topical), or tricyclic antidepressants within 3 days before the antivenom injection were excluded.
2.4. Measurements
To reconstitute the antivenom, 1 vial (20 mL) was diluted in 200–300 mL 0.9% sodium chloride (NaCl) and infused intravenously over 30 minutes. The antivenom treatments were FH, FN, and a combination of both types of antivenom (FH+FN). For skin test, 0.1 mL of 1:100 diluted antivenom was injected intradermally into the forearm. The positive skin test was defined as local redness or indurations > 5 mm in circumference at the injection site or any systemic allergic reactions within 30 minutes.9, 15 Patients who did not receive an antivenom skin test were allocated to the “not performed” group. The following patient demographics such as age, sex, triage, vital signs, underlying disease, patient disposition, hospital length of stay (LOS), and surgical interventions such as debridement, fasciotomy, and amputation were extracted from the CGRD database. Patients who developed skin rashes, redness of the eyes, eyelid swelling, itching, sneezing, wheezing, coughing, or bronchospasm within 3 hours after antivenom administration and received antihistamine, steroid, or bronchodilator treatment were defined as having antivenom reactions.20, 21 Patients who developed 1) hypotension after antivenom administration and were treated with fluid resuscitation or epinephrine injection or 2) respiratory failure due to the allergic reaction (neurotoxin-related respiratory failure was excluded) after antivenom administration with intubation were defined as experiencing severe reactions.19, 20
2.5 Data Analysis For continuous variables, age was summarized as means ± standard deviations. Medians with quartile deviation (QD) were used to present the doses of medications and the hospital LOS. The
distributions of categorical data were presented as numbers and percentages. The Student’s t-test, one-way analysis of variance (ANOVA), Mann–Whitney U test, and chi-square tests were used for the analyses. To determine the associations between the variables and antivenom reactions, binary logistic regression was used to adjust for potential confounding factors. The effects were estimated using adjusted odds ratios (aORs) and corresponding 95% confidence intervals (CIs). Results were considered statistically significant for a two-tailed test at a P value < 0.05. The IBM statistical package for the social sciences (SPSS) for Windows, version 22.0 (released 2013, IBM Corp., Armonk, NY) was used for all statistical analyses.
3. Results Patients who received an antihistamine or steroid within 3 days before the antivenom injection (n = 40) were excluded. This study enrolled 799 snakebite patients who were administered antivenom, including 373 and 26 who were subjected to a skin test and had negative and positive results, respectively, and 400 who did not undergo skin tests. In addition, 219 patients developed antivenom reactions, whereas 580 did not. Furthermore, 10 patients were identified to have experienced severe reactions (Figure 1). Among these 10 patients, 2 who experienced respiratory failure were intubated and injected with epinephrine, and 4 each experienced shock and were treated with epinephrine or hydration. Table 1 shows the demographic data. The skin test results and antivenom type and dose were associated with antivenom reactions. Patients with high-dose antivenom and a positive skin test result were associated with antivenom reactions (Figure 2). Patients who received combined
therapy (FH+FN), received a higher dose and showed higher incidence of antivenom reactions than those treated with single agents (Table 2). After adjusting for potential confounding factors using logistic regression, co-therapy (FH+FN), high dose (≥ 5 vials), and positive skin test were still associated with antivenom reactions (Table 3). To determine the antivenom skin test accuracy, its sensitivity, specificity, and positive and negative predictive values were analyzed, and are presented in Table 4. The antivenom skin test showed high specificity (98.5%, 95% CI: 97.49%–99.83%) but low sensitivity (17.5%, 95% CI: 10.74%–24.18%).
4. Discussion The antivenom reaction incidence in individuals administered snake antivenom varies among different studies,17-19, 22, 23 and reactions to horse-derived F(ab')2 antivenom range from 3% to 88%.14 In Taiwan, the difference still existed when Mao et al.17 reported that the antivenom reaction induction rate of FN was 22.9%, and Lin et al.22 reported rates of 4.7% and 7.0% in FH- and FN-treated groups respectively. In this study, the antivenom reaction rates were 24.4% and 20.0% in the FH and FN groups, respectively, and the difference might have been caused by differences in doses. While Lin et al.22 reported that most patients were administered one vial of antivenom, only 22.2% patients were administered one vial in this present study. Our study results indicate that antivenom reactions were related to co-treatment with different antivenoms and high doses and were associated with positive skin test results. A previous study
reported that patients co-administered different antivenoms had higher antivenom reaction rates than those administered a single type of antivenom did.22 Physicians tend to combine antivenom agents mainly because it is difficult to identify the snake species based on limited information provided by patients. Moreover, bites of N. atra in Taiwan mainly cause wound infection and tissue necrosis and not neurotoxicity.18 Consequently, physicians find it difficult to distinguish N. atra envenomation from that of other common snakes with hemorrhagic venom such as the green habu and Taiwan habu based on neurotoxic symptoms. For patients bitten by unknown snakes, physicians might prescribe antivenom combinations, and up to 8.6% of patients were administered two or more types in previously reported studies.22, 24 Considering the potential for antivenom reaction, combined antivenom therapy should not be recommended for routine treatment of snakebite victims. Physicians could delay antivenom therapy and monitor clinical changes, especially in those without obvious toxidrome. Distinguishing snake species according to distribution, local epidemiology, prevalence, location, affected body part, laboratory data, and wound type may also be helpful.16, 25, 26 High-dose antivenom treatment is also related to antivenom reactions.27 However, the pathophysiology of early antivenom reactions is currently unknown.28 In this study, patients who received ≥ 5 vials of antivenom had a higher risk of developing antivenom reactions than those who received fewer vials did. According to previous studies, the median dose is 10 and 7 vials for N. atra, and Bungarus multicinctus bites, respectively.17, 18 Patients who are bitten by N. atra and B. multicinctus and have to receive antivenom are at a higher risk of antivenom reactions than other
antivenom-treated patients are. In administering an antivenom, physicians should consider not only the toxidrome of the snake bite but also the potential allergic reaction. Theoretically, a skin test could be used to predict susceptibility to reactions in patients administered antivenom.29 In fact, some physicians perform the antivenom skin test according to the manufacturer’s instructions in many Asian countries. 18, 19, 30 However, regardless of the results of the skin test, physicians should use antivenoms to treat snakebites. Some studies suggest that the antivenom serum skin test is not reliable, 15, 16, 19 however, no study has reported the sensitivity and specificity. This study proved that patients with a positive skin test result had a higher antivenom reaction rate. Although the antivenom skin test was highly specific, the sensitivity was very low, which may be attributable to the dose. Furthermore, limiting the dose in the skin test might cause false negative results. Increasing the dose of the skin test might improve sensitivity, but according to this study, the antivenom reaction rate was obviously increased with administration of ≥ 5 vials of antivenom. It is difficult to conduct a skin test using such a high dose, so further studies to determine adequate skin test dosage might be necessary. Physicians should be careful in interpreting skin test results and be knowledgeable about the limitations of this examination. This study has some limitations that are worth mentioning. First, some patients are directly reactive to snake venom rather than the antivenom, and therefore, the study may have overestimated the antivenom reaction incidence. Second, a skin test could show false positive results because of irritation caused by antivenom or eczematous skin, because false negative results due to late-phase and non-immunological reactions are not detectable during observation for any early reactions
(which are IgE mediated) following a skin test. Furthermore, there was no positive or negative control group. Third, this study did not include data of reactions to antivenom against D. acutus and D. siamensis, because these antivenom were not available in the study units. The skin test accuracy of antivenom against D. acutus and D. siamensis as well as its reaction rate may need to be further analyzed. Fourth, some delayed-onset allergy reactions, which occur later may be caused by other drugs such as antibiotics, which are often given after bites of Naja atra, and serum sickness, which could occur in 5–14 days, were not discussed in this study. Fifth, the limitations of the retrospective design might have introduced some confounding factors that could cause antivenom reactions, which were not measured in this analysis. For example, patients who received medication from other medical units, which might influence allergic reactions or who had a history of horse serum allergy might also have been included in the analysis. Finally, there might be some variations in concentrations of the antivenom diluted solution and injection rate among patients, which could also influence the result. For example, some patients who had mild reactions might have had their antivenom infusion temporarily halted and then resumed at a slower rate, with no further reactions. Patients who were not treated with medications could not be identified in our data. All these factors could influence the result.
5. Conclusion In conclusion, snake antivenom reactions were related to combined antivenom treatment, high antivenom dose, and associated with positive skin test results. The skin test results showed high
specificity and low sensitivity. Finally, the antivenom skin test should be abolished because the sensitivity is extremely low and the result could be misinterpreted because of the limitation of this examination.
CRediT author statement Po-Chun Chuang: Conceptualization, Formal analysis, Data Curation, Writing - Original Draft, Writing - Review & Editing, Kang-Wei Chang: Data Curation Fu-Jen Cheng: Resources, Validation, Supervision Meng-Huan Wu: Resources, Validation, Supervision Ming-Ta Tsai: Resources, Validation Chao-Jui Li: Methodology, Formal analysis, Resources, Writing - Review & Editing, Supervision
Acknowledgments: none
Conflicts of interest statement: none to declare
Funding: This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Reference 1.
Gutierrez JM, Calvete JJ, Habib AG, Harrison RA, Williams DJ and Warrell DA. Snakebite
envenoming. Nat Rev Dis Primers. 2017;3:17063. 2.
Chippaux JP. Snake-bites: appraisal of the global situation. Bull World Health Organ.
1998;76:515-24. 3.
Alirol E, Sharma SK, Bawaskar HS, Kuch U and Chappuis F. Snake bite in South Asia: a
review. PLoS Negl Trop Dis. 2010;4:e603. 4.
Liu CC, You CH, Wang PJ, Yu JS, Huang GJ, Liu CH, Hsieh WC and Lin CC. Analysis of the
efficacy of Taiwanese freeze-dried neurotoxic antivenom against Naja kaouthia, Naja siamensis and Ophiophagus hannah through proteomics and animal model approaches. PLoS Negl Trop Dis. 2017;11:e0006138. 5.
Villalta M, Pla D, Yang SL, Sanz L, Segura A, Vargas M, Chen PY, Herrera M, Estrada R,
Cheng YF, Lee CD, Cerdas M, Chiang JR, Angulo Y, Leon G, Calvete JJ and Gutierrez JM. Snake venomics and antivenomics of Protobothrops mucrosquamatus and Viridovipera stejnegeri from Taiwan: keys to understand the variable immune response in horses. J Proteomics. 2012;75:5628-45. 6.
Huang RJ, Chen SW, Chen TK and Liau MY. [The detoxification of Naja naja atra venom and
preparation of potent antivenin]. Zhonghua Min Guo Wei Sheng Wu Ji Mian Yi Xue Za Zhi. 1985;18:177-83. 7.
Lee CH, Lee YC, Liang MH, Leu SJ, Lin LT, Chiang JR and Yang YY. Antibodies against
Venom of the Snake Deinagkistrodon acutus. Appl Environ Microbiol. 2016;82:71-80. 8.
Lee CH, Lee YC, Lee YL, Leu SJ, Lin LT, Chen CC, Chiang JR, Mwale PF, Tsai BY, Hung CS
and Yang YY. Single Chain Antibody Fragment against Venom from the Snake Daboia russelii formosensis. Toxins (Basel). 2017;9. 9.
https://www.fda.gov.tw/MLMS/ShowFile.aspx?LicId=23000001&Seq=001&Type=9.
10. Liu CH, Li CJ, Li J, Hsu CL, Chen CH, Chang HC and Xie WQ. Application research of snake venom protein and anti-venom serum manufacturing technology. Taiwan Epidemiology Bulletin. 2015;31:76-85. 11. de Silva HA, Ryan NM and de Silva HJ. Adverse reactions to snake antivenom, and their prevention and treatment. Br J Clin Pharmacol. 2016;81:446-52. 12. Gold BS, Dart RC and Barish RA. Bites of venomous snakes. N Engl J Med. 2002;347:347-56. 13. Ryan NM, Downes MA and Isbister GK. Clinical features of serum sickness after Australian snake antivenom. Toxicon. 2015;108:181-3. 14. Leon G, Herrera M, Segura A, Villalta M, Vargas M and Gutierrez JM. Pathogenic mechanisms underlying adverse reactions induced by intravenous administration of snake antivenoms. Toxicon. 2013;76:63-76. 15. Chen JC, Bullard MJ, Chiu TF, Ng CJ and Liaw SJ. Risk of immediate effects from F(ab)2 bivalent antivenin in Taiwan. Wilderness Environ Med. 2000;11:163-7. 16. Mao Y-C and Hung D-Z. Management of snake envenomation in Taiwan. Clinical Toxinology: Clinical Toxinology. 2013:44.
17. Mao YC, Liu PY, Chiang LC, Liao SC, Su HY, Hsieh SY and Yang CC. Bungarus multicinctus multicinctus Snakebite in Taiwan. Am J Trop Med Hyg. 2017;96:1497-1504. 18. Mao YC, Liu PY, Chiang LC, Lai CS, Lai KL, Ho CH, Wang TH and Yang CC. Naja atra snakebite in Taiwan. Clin Toxicol (Phila). 2018;56:273-280. 19. Thiansookon A and Rojnuckarin P. Low incidence of early reactions to horse-derived F(ab')(2) antivenom for snakebites in Thailand. Acta Trop. 2008;105:203-5. 20. Warrell DA. Guidelines for the management of snake-bites. World Health Organization. 2010. 21. Sampson HA, Munoz-Furlong A, Campbell RL, Adkinson NF, Jr., Bock SA, Branum A, Brown SG, Camargo CA, Jr., Cydulka R, Galli SJ, Gidudu J, Gruchalla RS, Harlor AD, Jr., Hepner DL, Lewis LM, Lieberman PL, Metcalfe DD, O'Connor R, Muraro A, Rudman A, Schmitt C, Scherrer D, Simons FE, Thomas S, Wood JP and Decker WW. Second symposium on the definition and management of anaphylaxis: summary report--Second National Institute of Allergy and Infectious Disease/Food Allergy and Anaphylaxis Network symposium. J Allergy Clin Immunol. 2006;117:391-7. 22. Lin CC, Chaou CH and Tseng CY. An investigation of snakebite antivenom usage in Taiwan. J Formos Med Assoc. 2016;115:672-7. 23. Ariaratnam CA, Sjostrom L, Raziek Z, Kularatne SA, Arachchi RW, Sheriff MH, Theakston RD and Warrell DA. An open, randomized comparative trial of two antivenoms for the treatment of envenoming by Sri Lankan Russell's viper (Daboia russelii russelii). Trans R Soc Trop Med Hyg. 2001;95:74-80.
24. Liu CH and Xie WQ. Analysis of the use of antivenom in 2008–2012 by the health insurance database. Taiwan Epidemiology Bulletin. 2017. 25. Wang J, Tsan Y, Yan-Chiao M and Wang L. Venomous snakebites and antivenom treatment according to a protocol for pediatric patients in Taiwan. Journal of Venomous Animals and Toxins including Tropical Diseases. 2009;15:667-679. 26. Hung DZ. Taiwan's venomous snakebite: epidemiological, evolution and geographic differences. Trans R Soc Trop Med Hyg. 2004;98:96-101. 27. Malasit P, Warrell DA, Chanthavanich P, Viravan C, Mongkolsapaya J, Singhthong B and Supich C. Prediction, prevention, and mechanism of early (anaphylactic) antivenom reactions in victims of snake bites. Br Med J (Clin Res Ed). 1986;292:17-20. 28. Stone SF, Isbister GK, Shahmy S, Mohamed F, Abeysinghe C, Karunathilake H, Ariaratnam A, Jacoby-Alner TE, Cotterell CL and Brown SG. Immune response to snake envenoming and treatment with antivenom; complement activation, cytokine production and mast cell degranulation. PLoS Negl Trop Dis. 2013;7:e2326. 29. Klaewsongkram J. A role of snake antivenom skin test from the allergist's point of view. Acta Trop. 2009;109:84-5; author reply 86. 30. Rha JH, Kwon SM, Oh JR, Han BK, Lee KH and Kim JH. Snakebite in Korea: A Guideline to Primary Surgical Management. Yonsei Med J. 2015;56:1443-8.
Table 1. Clinical characteristics of patients with snakebites in antivenom reaction and no reaction groups Reaction n = 219
No reaction n = 580
114 (52.1) 28 (12.8) 77 (35.2)
354 (61.0) 112 (19.3) 114 (19.7)
Antivenom dose 1–2 vials (n = 422) 3–4 vials (n = 200) ≥ 5 vials (n = 177) Antivenom skin tests results
102 (46.6) 56 (25.6) 61 (27.9)
320 (55.2) 144 (24.8) 116 (20.0)
Negative Positive Not performed Age
104 (47.5) 22 (10.0) 93 (42.5) 52.1 ± 17.7
269 (46.4) 4 (0.7) 307 (52.9) 52.6 ± 17.8
Male sex Diabetes mellitus End stage renal disease Liver cirrhosis Urgent triage (triage I and II)
152 (69.4) 5 (2.3) 1 (0.5) 0 (0.0) 86 (39.3)
407 (70.2) 41 (7.1) 8 (1.4) 3 (0.5) 2220 (37.9)
0.863 0.010 0.457 0.566 0.754
Heart rate during triage Mean arterial pressure during triage Length of stay, hours Admission ICU admission
88.0 ± 12.0 111.3 ± 11.5 55.0 ± 61.5 98 (44.7) 4 (1.8)
87.0 ± 12.0 111.3 ± 12.7 43.0 ± 66.5 260 (44.8) 13 (2.2)
0.404 0.528 0.143 1 1
Surgical intervention Debridement Fasciotomy Amputation
14 (6.4) 7 (3.2) 13 (5.9) 0 (0.0)
56 (9.7) 36 (6.2) 41 (7.1) 5 (0.9)
0.162 0.113 0.638 0.330
Antivenom types Freeze-dried Hemorrhagic (FH) Freeze-dried Neurotoxic (FN) Combined (FH + FN)
P-value
<0.001
0.037
<0.001
0.710
Data were presented as a number (percentage), Mean ± SD, or median ± quartile deviation; ICU, intensive care unit
Table 2. Antivenom doses used and reactions to different types FH group n = 468
FN group n= 140
FH+FN group n = 191
Antivenom dose 1–2 vials (n = 422) 3–4 vials (n = 200) ≥ 5 vials (n = 177)
2 ± 1.5
3 ± 1.5
3 ± 1.5
0.06
259 (53.3) 115 (24.6) 94 (20.1)
68 (48.6) 38 (27.1) 34 (24.3)
95 (49.7) 47 (24.6) 49 (25.7)
0.415
Antivenom reactions
114 (24.4)
28 (20.0)
77 (40.3)
<0.001
Data were presented as number (percentage) or median ± quartile deviation
P-value
Table 3. Association of antivenom reactions with antivenom types, dose, and skin test results aOR
95% CI
Freeze-dried Hemorrhagic (FH) Freeze-dried Neurotoxic (FN) Combined (FH+FN) Antivenom Dose
0.5 0.4 1 (Reference)
0.35 to 0.74 0.24 to 0.69
1–2 vials 3–4 vials ≥ 5 vials Antivenom skin test results Negative
1 (Reference) 1.3 1.8
Antivenom types
Positive Not performed
0.85 to 1.90 1.23 to 2.76
1 (Reference) 16.7 0.9
5.42 to 51.22 0.63 to 1.25
aOR, adjusted odds ratio; CI, confidence interval. Confounders the model was adjusted for: age, male sex, diabetes milletus (DM), end-stage renal disease (ESRD), and liver cirrhosis (LC).
Table 4. Sensitivity, specificity, and positive/negative predictive value of antivenom skin test To predict reaction
To prediect severe antivenom reaction
Reaction n = 126
Severe reaction n=6
No reaction n = 273
No severe reaction n = 393
No. of positive skin tests No. of negative skin tests Sensitivity, % (95% CI)
22 4 104 269 17.5% (10.74% to 24.18%)
1 25 5 368 16.7% (-26.18% to 59.51%)
Specificity, % (95% CI) Positive predictive value, % Negative predicitve value, %
98.5% (97.10% to 99.97%) 84.6% (69.75% to 99.48%) 72.1% (67.55% to 76.69%)
93.6% (91.22% to 96.06%) 3.8% (-4.08% to 11.77%) 98.7% (97.49% to 99.83%)
N = 399
Figure Legends
Patients included N=799
Skin test negative n=373
No reaction n=269
Reaction n=104
Skin test positive n=26
No reaction n=4
Severe reaction n=5
Reaction n=22
Severe reaction n=1
Did not undergo skin test n=400
No reaction n=307
Reaction n=93
Severe reaction n=4
Figure 1. Flowchart. Patients who received any class of antihistamines (including H1 or H2 blockers), steroids (including systemic or topical), or tricyclic antidepressants within 3 days before the antivenom injection were not included in study.
a) Antivenom dose 1-2
3-4
b) Skin test results ≥5
Negative
Positive
100%
Antivenom reaction rate
90%
84.6%
80% 70% 60% 50% 40% 30%
34.5% 24.2%
28.0%
27.9%
20% 10% 0% p=0.035
p<0.001
Figure 2. Antivenom reaction rate of a) different doses and b) negative and positive skin test groups.
Risk factors associated with snake antivenom reaction and the role of skin test
Po-Chun Chuang ConceptualizationFormal analysisData CurationWriting - Original DraftWriting - Review & Editing Kang-Wei Chang Data Curation , Fu-Jen Cheng ResourcesValidationSupervision , Meng-Huan Wu ResourcesValidationSupervision , Ming-Ta Tsai ResourcesValidation , Chao-Jui Li MethodologyFormal analysisResourcesWriting - Review & EditingSupervision PII: DOI: Reference:
S0001-706X(19)31445-7 https://doi.org/10.1016/j.actatropica.2019.105293 ACTROP 105293
To appear in:
Acta Tropica
Received date: Revised date: Accepted date:
18 October 2019 22 November 2019 6 December 2019
Please cite this article as: Po-Chun Chuang ConceptualizationFormal analysisData CurationWriting - Original DraftW Kang-Wei Chang Data Curation , Fu-Jen Cheng ResourcesValidationSupervision , Meng-Huan Wu ResourcesValidationSupervision , Ming-Ta Tsai ResourcesValidation , Chao-Jui Li MethodologyFormal analysisResourcesWriting - Review & EditingSupervision , Risk factors associated with snake antivenom reaction and the role of skin test, Acta Tropica (2019), doi: https://doi.org/10.1016/j.actatropica.2019.105293
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Published by Elsevier B.V.
Risk factors associated with snake antivenom reaction and the role of skin test
Po-Chun Chuang, Kang-Wei Chang, Fu-Jen Cheng, Meng-Huan Wu, Ming-Ta Tsai, and Chao-Jui Li*
Department of Emergency Medicine, Kaohsiung Chang Gung Memorial Hospital, Chang Gung University College of Medicine, No. 123, Dapi Rd., Niaosong Dist., Kaohsiung City 833, Taiwan (R.O.C.)
*Corresponding author: Chao-Jui Li, Department of Emergency Medicine, Kaohsiung Chang Gung Memorial Hospital, Chang Gung University College of Medicine, No. 123, Dapi Rd., Niaosong Dist., Kaohsiung City 833, Taiwan (R.O.C.) Tel.: +886-912-188-889; Fax: +886-7-7317123-8415; Email: [email protected]
Abstract Antivenom reactions are a common complication of snake antivenom. This study aimed to identify predicators of antivenom reaction and the involvement of antivenom skin test in antivenom reaction development. This retrospective cohort study was conducted in six medical institutions in Taiwan. Data were extracted from the Chang Gung Research Database (CGRD) from January 2006 to December 2016. The association between antivenom reaction and patient demographics, type and dose of antivenom, and skin test results was analyzed. The study enrolled 799 patients, including 219 who developed antivenom reactions. Compared to patients receiving both freeze-dried hemorrhagic (FH) and freeze-dried neurotoxic (FN) antivenom, those administered a single type had a lower antivenom reaction risk (adjusted odds ratios [aORs]: 0.5 and 0.4, 95% confidence interval [CI]: 0.35–0.74 and 0.24–0.69, FH and FN respectively). Patients administered a higher antivenom dose (≥ 5 vials) had higher antivenom reaction risk (aOR: 1.8, 95% CI: 1.23–2.76). A positive skin test result was also associated with antivenom reaction (aOR: 16.7, 95% CI: 5.42– 51.22). The skin test showed high specificity (98.5%, 95% CI: 97.49%–99.83%) but low sensitivity (17.5%, 95% CI: 10.74%–24.18%). The antivenom skin test should be abolished because of the extremely low sensitivity and possible misinterpretation of results because of the limitation of this examination.
Keywords: snakes, antivenins, allergy, antivenom reactions, skin test, Taiwan
1. Introduction Snakebite envenomation is a tropical disease that causes mortality and morbidity in more than 100,000 and over 400,000 people, respectively yearly.1-3 Taiwan is located in Southeast Asia and has approximately 23 species of snakes, including 6 common clinically significant species. 4 Snake venom toxicity can be divided into hemorrhagic including that caused by the green habu (Viridovipera stejnegeri), Taiwan habu (Protobothrops mucrosquamatus), and hundred-pacer (Deinagkistrodon acutus) snakes, and neurotoxic, which includes toxicity caused by the Taiwan banded krait (Bungarus multicinctus) and the Chinese cobra (Naja atra), while the Russell’s viper (Daboia siamensis) has mixed hemorrhagic and neurotoxic venom. Treatment of snake bites mainly involves the injection of horse immunoglobulin (Ig) antivenom. Horse Ig antivenoms produced by the Centers for Disease Control, R.O.C (Taiwan) are freeze-dried hemorrhagic (FH) bivalent antivenom against the hemorrhagic venom of V. stejnegeri and P. mucrosquamatus; freeze-dried neurotoxic (FN), a bivalent antivenom against the neurotoxic venom of B. multicinctus and N. atra; freeze-dried acutus (FA), an antivenom against the hemorrhagic venom of D. acutus; and freeze-dried D. russellii (FR), which was developed in 2008 against the mixed-type venom of D. siamensis.4-8 These antivenoms are vacuum freeze-dried F(abʹ)2 fragments from horses, which contain more than 1000 units of antivenom serum; and the excipient contains the preservative (Thiomersal).9 Combining F(abʹ)2 fragments, which are 100 kDa with a half-life of approximately 2–4 days, with snake venom antigens forms stable and easily cleared immune complexes.10
Snake antivenoms can cause reactions ranging from mild skin rash to life-threatening anaphylaxis.11 Early reactions occur soon after antivenom injection (often within 1 h)11, 12, whereas late (serum sickness) reactions could develop between 5 and 14 days.11, 13 Early reactions could be due to the combination of type I (IgE) hypersensitivity, immune complex aggregation, and activation of the complement.11, 14 Late (serum sickness) reactions are mediated by type III (IgG or IgM) hypersensitivity.14 To predict antivenom reactions, the current treatment principle and manufacturer instructions of antivenom in Taiwan suggest that a skin test should be conducted before administration.9,
15, 16
However, the effects of skin tests are unclear, especially because
clinically regardless of the results of the skin test, physicians should use antivenoms to treat snakebites.17, 18 A previous case series study reported that the skin test was not helpful in predicting antivenom reactions19, and its sensitivity and specificity were unclear. Furthermore, the World Health Organization (WHO) guideline suggests that an antivenom skin test should not be used.20 Therefore, this study aimed to analyze the skin test accuracy and risk factors associated with antivenom reaction.
2. Methods 2.1 Ethics This retrospective study was approved by the Chang Gung Medical Foundation Institutional Review Board (IRB No.: 201801929B0). All patient data used in the analyses were anonymized.
2.2 Study setting The data used were obtained from the largest healthcare system in Taiwan, the Chang Gung Memorial Hospital (CGMH), which receives 11.5% of the National Health Insurance budget according to government statistics. The Chang Gung Research Database (CGRD), which combines original medical records from six medical institutions, Keelung, Taipei, Linkou, Yunlin, Chiayi, and Kaohsiung branches located from northern to southern Taiwan, was used.
2.3 Patients All patients who experienced snake bites, visited the emergency department (ED), and received antivenom injections from January 2006 to December 2016 were included in the study. Patients who received any class of antihistamine (including H1 or H2 blockers), steroid (including systemic or topical), or tricyclic antidepressants within 3 days before the antivenom injection were excluded.
2.4. Measurements
To reconstitute the antivenom, 1 vial (20 mL) was diluted in 200–300 mL 0.9% sodium chloride (NaCl) and infused intravenously over 30 minutes. The antivenom treatments were FH, FN, and a combination of both types of antivenom (FH+FN). For skin test, 0.1 mL of 1:100 diluted antivenom was injected intradermally into the forearm. The positive skin test was defined as local redness or indurations > 5 mm in circumference at the injection site or any systemic allergic reactions within 30 minutes.9, 15 Patients who did not receive an antivenom skin test were allocated to the “not performed” group. The following patient demographics such as age, sex, triage, vital signs, underlying disease, patient disposition, hospital length of stay (LOS), and surgical interventions such as debridement, fasciotomy, and amputation were extracted from the CGRD database. Patients who developed skin rashes, redness of the eyes, eyelid swelling, itching, sneezing, wheezing, coughing, or bronchospasm within 3 hours after antivenom administration and received antihistamine, steroid, or bronchodilator treatment were defined as having antivenom reactions.20, 21 Patients who developed 1) hypotension after antivenom administration and were treated with fluid resuscitation or epinephrine injection or 2) respiratory failure due to the allergic reaction (neurotoxin-related respiratory failure was excluded) after antivenom administration with intubation were defined as experiencing severe reactions.19, 20
2.5 Data Analysis For continuous variables, age was summarized as means ± standard deviations. Medians with quartile deviation (QD) were used to present the doses of medications and the hospital LOS. The
distributions of categorical data were presented as numbers and percentages. The Student’s t-test, one-way analysis of variance (ANOVA), Mann–Whitney U test, and chi-square tests were used for the analyses. To determine the associations between the variables and antivenom reactions, binary logistic regression was used to adjust for potential confounding factors. The effects were estimated using adjusted odds ratios (aORs) and corresponding 95% confidence intervals (CIs). Results were considered statistically significant for a two-tailed test at a P value < 0.05. The IBM statistical package for the social sciences (SPSS) for Windows, version 22.0 (released 2013, IBM Corp., Armonk, NY) was used for all statistical analyses.
3. Results Patients who received an antihistamine or steroid within 3 days before the antivenom injection (n = 40) were excluded. This study enrolled 799 snakebite patients who were administered antivenom, including 373 and 26 who were subjected to a skin test and had negative and positive results, respectively, and 400 who did not undergo skin tests. In addition, 219 patients developed antivenom reactions, whereas 580 did not. Furthermore, 10 patients were identified to have experienced severe reactions (Figure 1). Among these 10 patients, 2 who experienced respiratory failure were intubated and injected with epinephrine, and 4 each experienced shock and were treated with epinephrine or hydration. Table 1 shows the demographic data. The skin test results and antivenom type and dose were associated with antivenom reactions. Patients with high-dose antivenom and a positive skin test result were associated with antivenom reactions (Figure 2). Patients who received combined
therapy (FH+FN), received a higher dose and showed higher incidence of antivenom reactions than those treated with single agents (Table 2). After adjusting for potential confounding factors using logistic regression, co-therapy (FH+FN), high dose (≥ 5 vials), and positive skin test were still associated with antivenom reactions (Table 3). To determine the antivenom skin test accuracy, its sensitivity, specificity, and positive and negative predictive values were analyzed, and are presented in Table 4. The antivenom skin test showed high specificity (98.5%, 95% CI: 97.49%–99.83%) but low sensitivity (17.5%, 95% CI: 10.74%–24.18%).
4. Discussion The antivenom reaction incidence in individuals administered snake antivenom varies among different studies,17-19, 22, 23 and reactions to horse-derived F(ab')2 antivenom range from 3% to 88%.14 In Taiwan, the difference still existed when Mao et al.17 reported that the antivenom reaction induction rate of FN was 22.9%, and Lin et al.22 reported rates of 4.7% and 7.0% in FH- and FN-treated groups respectively. In this study, the antivenom reaction rates were 24.4% and 20.0% in the FH and FN groups, respectively, and the difference might have been caused by differences in doses. While Lin et al.22 reported that most patients were administered one vial of antivenom, only 22.2% patients were administered one vial in this present study. Our study results indicate that antivenom reactions were related to co-treatment with different antivenoms and high doses and were associated with positive skin test results. A previous study
reported that patients co-administered different antivenoms had higher antivenom reaction rates than those administered a single type of antivenom did.22 Physicians tend to combine antivenom agents mainly because it is difficult to identify the snake species based on limited information provided by patients. Moreover, bites of N. atra in Taiwan mainly cause wound infection and tissue necrosis and not neurotoxicity.18 Consequently, physicians find it difficult to distinguish N. atra envenomation from that of other common snakes with hemorrhagic venom such as the green habu and Taiwan habu based on neurotoxic symptoms. For patients bitten by unknown snakes, physicians might prescribe antivenom combinations, and up to 8.6% of patients were administered two or more types in previously reported studies.22, 24 Considering the potential for antivenom reaction, combined antivenom therapy should not be recommended for routine treatment of snakebite victims. Physicians could delay antivenom therapy and monitor clinical changes, especially in those without obvious toxidrome. Distinguishing snake species according to distribution, local epidemiology, prevalence, location, affected body part, laboratory data, and wound type may also be helpful.16, 25, 26 High-dose antivenom treatment is also related to antivenom reactions.27 However, the pathophysiology of early antivenom reactions is currently unknown.28 In this study, patients who received ≥ 5 vials of antivenom had a higher risk of developing antivenom reactions than those who received fewer vials did. According to previous studies, the median dose is 10 and 7 vials for N. atra, and Bungarus multicinctus bites, respectively.17, 18 Patients who are bitten by N. atra and B. multicinctus and have to receive antivenom are at a higher risk of antivenom reactions than other
antivenom-treated patients are. In administering an antivenom, physicians should consider not only the toxidrome of the snake bite but also the potential allergic reaction. Theoretically, a skin test could be used to predict susceptibility to reactions in patients administered antivenom.29 In fact, some physicians perform the antivenom skin test according to the manufacturer’s instructions in many Asian countries. 18, 19, 30 However, regardless of the results of the skin test, physicians should use antivenoms to treat snakebites. Some studies suggest that the antivenom serum skin test is not reliable, 15, 16, 19 however, no study has reported the sensitivity and specificity. This study proved that patients with a positive skin test result had a higher antivenom reaction rate. Although the antivenom skin test was highly specific, the sensitivity was very low, which may be attributable to the dose. Furthermore, limiting the dose in the skin test might cause false negative results. Increasing the dose of the skin test might improve sensitivity, but according to this study, the antivenom reaction rate was obviously increased with administration of ≥ 5 vials of antivenom. It is difficult to conduct a skin test using such a high dose, so further studies to determine adequate skin test dosage might be necessary. Physicians should be careful in interpreting skin test results and be knowledgeable about the limitations of this examination. This study has some limitations that are worth mentioning. First, some patients are directly reactive to snake venom rather than the antivenom, and therefore, the study may have overestimated the antivenom reaction incidence. Second, a skin test could show false positive results because of irritation caused by antivenom or eczematous skin, because false negative results due to late-phase and non-immunological reactions are not detectable during observation for any early reactions
(which are IgE mediated) following a skin test. Furthermore, there was no positive or negative control group. Third, this study did not include data of reactions to antivenom against D. acutus and D. siamensis, because these antivenom were not available in the study units. The skin test accuracy of antivenom against D. acutus and D. siamensis as well as its reaction rate may need to be further analyzed. Fourth, some delayed-onset allergy reactions, which occur later may be caused by other drugs such as antibiotics, which are often given after bites of Naja atra, and serum sickness, which could occur in 5–14 days, were not discussed in this study. Fifth, the limitations of the retrospective design might have introduced some confounding factors that could cause antivenom reactions, which were not measured in this analysis. For example, patients who received medication from other medical units, which might influence allergic reactions or who had a history of horse serum allergy might also have been included in the analysis. Finally, there might be some variations in concentrations of the antivenom diluted solution and injection rate among patients, which could also influence the result. For example, some patients who had mild reactions might have had their antivenom infusion temporarily halted and then resumed at a slower rate, with no further reactions. Patients who were not treated with medications could not be identified in our data. All these factors could influence the result.
5. Conclusion In conclusion, snake antivenom reactions were related to combined antivenom treatment, high antivenom dose, and associated with positive skin test results. The skin test results showed high
specificity and low sensitivity. Finally, the antivenom skin test should be abolished because the sensitivity is extremely low and the result could be misinterpreted because of the limitation of this examination.
CRediT author statement Po-Chun Chuang: Conceptualization, Formal analysis, Data Curation, Writing - Original Draft, Writing - Review & Editing, Kang-Wei Chang: Data Curation Fu-Jen Cheng: Resources, Validation, Supervision Meng-Huan Wu: Resources, Validation, Supervision Ming-Ta Tsai: Resources, Validation Chao-Jui Li: Methodology, Formal analysis, Resources, Writing - Review & Editing, Supervision
Acknowledgments: none
Conflicts of interest statement: none to declare
Funding: This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Reference 1.
Gutierrez JM, Calvete JJ, Habib AG, Harrison RA, Williams DJ and Warrell DA. Snakebite
envenoming. Nat Rev Dis Primers. 2017;3:17063. 2.
Chippaux JP. Snake-bites: appraisal of the global situation. Bull World Health Organ.
1998;76:515-24. 3.
Alirol E, Sharma SK, Bawaskar HS, Kuch U and Chappuis F. Snake bite in South Asia: a
review. PLoS Negl Trop Dis. 2010;4:e603. 4.
Liu CC, You CH, Wang PJ, Yu JS, Huang GJ, Liu CH, Hsieh WC and Lin CC. Analysis of the
efficacy of Taiwanese freeze-dried neurotoxic antivenom against Naja kaouthia, Naja siamensis and Ophiophagus hannah through proteomics and animal model approaches. PLoS Negl Trop Dis. 2017;11:e0006138. 5.
Villalta M, Pla D, Yang SL, Sanz L, Segura A, Vargas M, Chen PY, Herrera M, Estrada R,
Cheng YF, Lee CD, Cerdas M, Chiang JR, Angulo Y, Leon G, Calvete JJ and Gutierrez JM. Snake venomics and antivenomics of Protobothrops mucrosquamatus and Viridovipera stejnegeri from Taiwan: keys to understand the variable immune response in horses. J Proteomics. 2012;75:5628-45. 6.
Huang RJ, Chen SW, Chen TK and Liau MY. [The detoxification of Naja naja atra venom and
preparation of potent antivenin]. Zhonghua Min Guo Wei Sheng Wu Ji Mian Yi Xue Za Zhi. 1985;18:177-83. 7.
Lee CH, Lee YC, Liang MH, Leu SJ, Lin LT, Chiang JR and Yang YY. Antibodies against
Venom of the Snake Deinagkistrodon acutus. Appl Environ Microbiol. 2016;82:71-80. 8.
Lee CH, Lee YC, Lee YL, Leu SJ, Lin LT, Chen CC, Chiang JR, Mwale PF, Tsai BY, Hung CS
and Yang YY. Single Chain Antibody Fragment against Venom from the Snake Daboia russelii formosensis. Toxins (Basel). 2017;9. 9.
https://www.fda.gov.tw/MLMS/ShowFile.aspx?LicId=23000001&Seq=001&Type=9.
10. Liu CH, Li CJ, Li J, Hsu CL, Chen CH, Chang HC and Xie WQ. Application research of snake venom protein and anti-venom serum manufacturing technology. Taiwan Epidemiology Bulletin. 2015;31:76-85. 11. de Silva HA, Ryan NM and de Silva HJ. Adverse reactions to snake antivenom, and their prevention and treatment. Br J Clin Pharmacol. 2016;81:446-52. 12. Gold BS, Dart RC and Barish RA. Bites of venomous snakes. N Engl J Med. 2002;347:347-56. 13. Ryan NM, Downes MA and Isbister GK. Clinical features of serum sickness after Australian snake antivenom. Toxicon. 2015;108:181-3. 14. Leon G, Herrera M, Segura A, Villalta M, Vargas M and Gutierrez JM. Pathogenic mechanisms underlying adverse reactions induced by intravenous administration of snake antivenoms. Toxicon. 2013;76:63-76. 15. Chen JC, Bullard MJ, Chiu TF, Ng CJ and Liaw SJ. Risk of immediate effects from F(ab)2 bivalent antivenin in Taiwan. Wilderness Environ Med. 2000;11:163-7. 16. Mao Y-C and Hung D-Z. Management of snake envenomation in Taiwan. Clinical Toxinology: Clinical Toxinology. 2013:44.
17. Mao YC, Liu PY, Chiang LC, Liao SC, Su HY, Hsieh SY and Yang CC. Bungarus multicinctus multicinctus Snakebite in Taiwan. Am J Trop Med Hyg. 2017;96:1497-1504. 18. Mao YC, Liu PY, Chiang LC, Lai CS, Lai KL, Ho CH, Wang TH and Yang CC. Naja atra snakebite in Taiwan. Clin Toxicol (Phila). 2018;56:273-280. 19. Thiansookon A and Rojnuckarin P. Low incidence of early reactions to horse-derived F(ab')(2) antivenom for snakebites in Thailand. Acta Trop. 2008;105:203-5. 20. Warrell DA. Guidelines for the management of snake-bites. World Health Organization. 2010. 21. Sampson HA, Munoz-Furlong A, Campbell RL, Adkinson NF, Jr., Bock SA, Branum A, Brown SG, Camargo CA, Jr., Cydulka R, Galli SJ, Gidudu J, Gruchalla RS, Harlor AD, Jr., Hepner DL, Lewis LM, Lieberman PL, Metcalfe DD, O'Connor R, Muraro A, Rudman A, Schmitt C, Scherrer D, Simons FE, Thomas S, Wood JP and Decker WW. Second symposium on the definition and management of anaphylaxis: summary report--Second National Institute of Allergy and Infectious Disease/Food Allergy and Anaphylaxis Network symposium. J Allergy Clin Immunol. 2006;117:391-7. 22. Lin CC, Chaou CH and Tseng CY. An investigation of snakebite antivenom usage in Taiwan. J Formos Med Assoc. 2016;115:672-7. 23. Ariaratnam CA, Sjostrom L, Raziek Z, Kularatne SA, Arachchi RW, Sheriff MH, Theakston RD and Warrell DA. An open, randomized comparative trial of two antivenoms for the treatment of envenoming by Sri Lankan Russell's viper (Daboia russelii russelii). Trans R Soc Trop Med Hyg. 2001;95:74-80.
24. Liu CH and Xie WQ. Analysis of the use of antivenom in 2008–2012 by the health insurance database. Taiwan Epidemiology Bulletin. 2017. 25. Wang J, Tsan Y, Yan-Chiao M and Wang L. Venomous snakebites and antivenom treatment according to a protocol for pediatric patients in Taiwan. Journal of Venomous Animals and Toxins including Tropical Diseases. 2009;15:667-679. 26. Hung DZ. Taiwan's venomous snakebite: epidemiological, evolution and geographic differences. Trans R Soc Trop Med Hyg. 2004;98:96-101. 27. Malasit P, Warrell DA, Chanthavanich P, Viravan C, Mongkolsapaya J, Singhthong B and Supich C. Prediction, prevention, and mechanism of early (anaphylactic) antivenom reactions in victims of snake bites. Br Med J (Clin Res Ed). 1986;292:17-20. 28. Stone SF, Isbister GK, Shahmy S, Mohamed F, Abeysinghe C, Karunathilake H, Ariaratnam A, Jacoby-Alner TE, Cotterell CL and Brown SG. Immune response to snake envenoming and treatment with antivenom; complement activation, cytokine production and mast cell degranulation. PLoS Negl Trop Dis. 2013;7:e2326. 29. Klaewsongkram J. A role of snake antivenom skin test from the allergist's point of view. Acta Trop. 2009;109:84-5; author reply 86. 30. Rha JH, Kwon SM, Oh JR, Han BK, Lee KH and Kim JH. Snakebite in Korea: A Guideline to Primary Surgical Management. Yonsei Med J. 2015;56:1443-8.
Table 1. Clinical characteristics of patients with snakebites in antivenom reaction and no reaction groups Reaction n = 219
No reaction n = 580
114 (52.1) 28 (12.8) 77 (35.2)
354 (61.0) 112 (19.3) 114 (19.7)
Antivenom dose 1–2 vials (n = 422) 3–4 vials (n = 200) ≥ 5 vials (n = 177) Antivenom skin tests results
102 (46.6) 56 (25.6) 61 (27.9)
320 (55.2) 144 (24.8) 116 (20.0)
Negative Positive Not performed Age
104 (47.5) 22 (10.0) 93 (42.5) 52.1 ± 17.7
269 (46.4) 4 (0.7) 307 (52.9) 52.6 ± 17.8
Male sex Diabetes mellitus End stage renal disease Liver cirrhosis Urgent triage (triage I and II)
152 (69.4) 5 (2.3) 1 (0.5) 0 (0.0) 86 (39.3)
407 (70.2) 41 (7.1) 8 (1.4) 3 (0.5) 2220 (37.9)
0.863 0.010 0.457 0.566 0.754
Heart rate during triage Mean arterial pressure during triage Length of stay, hours Admission ICU admission
88.0 ± 12.0 111.3 ± 11.5 55.0 ± 61.5 98 (44.7) 4 (1.8)
87.0 ± 12.0 111.3 ± 12.7 43.0 ± 66.5 260 (44.8) 13 (2.2)
0.404 0.528 0.143 1 1
Surgical intervention Debridement Fasciotomy Amputation
14 (6.4) 7 (3.2) 13 (5.9) 0 (0.0)
56 (9.7) 36 (6.2) 41 (7.1) 5 (0.9)
0.162 0.113 0.638 0.330
Antivenom types Freeze-dried Hemorrhagic (FH) Freeze-dried Neurotoxic (FN) Combined (FH + FN)
P-value
<0.001
0.037
<0.001
0.710
Data were presented as a number (percentage), Mean ± SD, or median ± quartile deviation; ICU, intensive care unit
Table 2. Antivenom doses used and reactions to different types FH group n = 468
FN group n= 140
FH+FN group n = 191
Antivenom dose 1–2 vials (n = 422) 3–4 vials (n = 200) ≥ 5 vials (n = 177)
2 ± 1.5
3 ± 1.5
3 ± 1.5
0.06
259 (53.3) 115 (24.6) 94 (20.1)
68 (48.6) 38 (27.1) 34 (24.3)
95 (49.7) 47 (24.6) 49 (25.7)
0.415
Antivenom reactions
114 (24.4)
28 (20.0)
77 (40.3)
<0.001
Data were presented as number (percentage) or median ± quartile deviation
P-value
Table 3. Association of antivenom reactions with antivenom types, dose, and skin test results aOR
95% CI
Freeze-dried Hemorrhagic (FH) Freeze-dried Neurotoxic (FN) Combined (FH+FN) Antivenom Dose
0.5 0.4 1 (Reference)
0.35 to 0.74 0.24 to 0.69
1–2 vials 3–4 vials ≥ 5 vials Antivenom skin test results Negative
1 (Reference) 1.3 1.8
Antivenom types
Positive Not performed
0.85 to 1.90 1.23 to 2.76
1 (Reference) 16.7 0.9
5.42 to 51.22 0.63 to 1.25
aOR, adjusted odds ratio; CI, confidence interval. Confounders the model was adjusted for: age, male sex, diabetes milletus (DM), end-stage renal disease (ESRD), and liver cirrhosis (LC).
Table 4. Sensitivity, specificity, and positive/negative predictive value of antivenom skin test To predict reaction
To prediect severe antivenom reaction
Reaction n = 126
Severe reaction n=6
No reaction n = 273
No severe reaction n = 393
No. of positive skin tests No. of negative skin tests Sensitivity, % (95% CI)
22 4 104 269 17.5% (10.74% to 24.18%)
1 25 5 368 16.7% (-26.18% to 59.51%)
Specificity, % (95% CI) Positive predictive value, % Negative predicitve value, %
98.5% (97.10% to 99.97%) 84.6% (69.75% to 99.48%) 72.1% (67.55% to 76.69%)
93.6% (91.22% to 96.06%) 3.8% (-4.08% to 11.77%) 98.7% (97.49% to 99.83%)
N = 399
Figure Legends
Patients included N=799
Skin test negative n=373
No reaction n=269
Reaction n=104
Skin test positive n=26
No reaction n=4
Severe reaction n=5
Reaction n=22
Severe reaction n=1
Did not undergo skin test n=400
No reaction n=307
Reaction n=93
Severe reaction n=4
Figure 1. Flowchart. Patients who received any class of antihistamines (including H1 or H2 blockers), steroids (including systemic or topical), or tricyclic antidepressants within 3 days before the antivenom injection were not included in study.
a) Antivenom dose 1-2
3-4
b) Skin test results ≥5
Negative
Positive
100%
Antivenom reaction rate
90%
84.6%
80% 70% 60% 50% 40% 30%
34.5% 24.2%
28.0%
27.9%
20% 10% 0% p=0.035
p<0.001
Figure 2. Antivenom reaction rate of a) different doses and b) negative and positive skin test groups.